<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
bert-base-uncased-finetuned-vr-comfort-2125
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0440
- Accuracy: 0.8431
- F1: 0.8437
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.7159 | 1.0 | 157 | 0.6408 | 0.7401 | 0.6612 |
| 0.5559 | 2.0 | 314 | 0.5362 | 0.7952 | 0.7684 |
| 0.362 | 3.0 | 471 | 0.5135 | 0.8204 | 0.8132 |
| 0.1918 | 4.0 | 628 | 0.6109 | 0.8407 | 0.8388 |
| 0.1192 | 5.0 | 785 | 0.6947 | 0.8347 | 0.8316 |
| 0.0661 | 6.0 | 942 | 0.7843 | 0.8455 | 0.8467 |
| 0.0507 | 7.0 | 1099 | 0.9312 | 0.8168 | 0.8271 |
| 0.0406 | 8.0 | 1256 | 0.8616 | 0.8467 | 0.8488 |
| 0.0268 | 9.0 | 1413 | 0.8403 | 0.8443 | 0.8478 |
| 0.0251 | 10.0 | 1570 | 0.8662 | 0.8467 | 0.8472 |
| 0.0188 | 11.0 | 1727 | 0.9418 | 0.8563 | 0.8530 |
| 0.0195 | 12.0 | 1884 | 0.9541 | 0.8479 | 0.8469 |
| 0.0172 | 13.0 | 2041 | 0.9372 | 0.8407 | 0.8413 |
| 0.0142 | 14.0 | 2198 | 0.9883 | 0.8491 | 0.8469 |
| 0.0156 | 15.0 | 2355 | 1.0150 | 0.8419 | 0.8428 |
| 0.0138 | 16.0 | 2512 | 1.0035 | 0.8479 | 0.8466 |
| 0.013 | 17.0 | 2669 | 1.0909 | 0.8299 | 0.8355 |
| 0.0115 | 18.0 | 2826 | 1.0278 | 0.8515 | 0.8490 |
| 0.0107 | 19.0 | 2983 | 1.0419 | 0.8431 | 0.8437 |
| 0.0101 | 20.0 | 3140 | 1.0440 | 0.8431 | 0.8437 |
Framework versions
- Transformers 4.13.0
- Pytorch 1.11.0
- Datasets 1.16.1
- Tokenizers 0.10.3
